Applied behavioral analysis (ABA) is widely considered the gold standard for autism treatment, but accessing ABA is easier said than done given the huge number of children on the spectrum. 

In fact, families are often forced onto treatment waitlists due to the nationwide shortage of ABA clinicians. And that’s if they can get an autism diagnosis at all, as diagnostic services themselves are also in short supply.

Parents who suspect their child has autism often turn to pediatricians for help. And from there, doctors typically refer children out to a specialist for diagnosis, leading to long, harmful delays in treatment.

But Cognoa, a digital pediatric behavioral health company, is trying to change that. It’s using artificial intelligence (AI) and machine learning to help empower pediatricians to make autism diagnoses themselves, getting children into treatment quicker and eliminating gender, racial, ethnic and socio-economic diagnosis disparities.

Since Cognoa’s founding in 2013, the Palo Alto, California-based company has raised nearly $54 million dollars, according to the fundraising tracker website Crunchbase. Plus, it’s currently on its way toward becoming the first FDA-cleared solution that enables autism diagnoses in the primary care setting.

Behavioral Health Business recently connected with CEO Dave Happel to discuss how Cognoa stands to impact autism treatment providers long-term, when it’s planning a widespread rollout of the tool and where the company is headed next.

You can find the conversation below, edited for length and clarity.  

Inside the C-Suite shines a spotlight on executives in the behavioral health space. Know a top leader who’d like to be profiled in an upcoming Q&A? Drop us a line at [email protected].

BHB: Can you tell me a little bit about Cognoa’s origin story and how the company got its start?

Happel: Cognoa was founded in 2013 by Dennis Wall, PhD, Associate Professor of Pediatrics and Psychiatry at Stanford University, with the aim of transforming the current system of care for children and families living with autism.

Currently, pediatricians refer most children with suspected developmental delays to specialists to diagnose and prescribe treatment. This often results in children and families facing a long, arduous journey before they are able to access life-changing therapy, typically waiting months or even years before receiving an initial autism diagnosis.

Dennis founded Cognoa with the goal of fundamentally changing this reality with an approach to enable physicians to make a diagnosis in the primary care setting. Parents tend to express concern about behavioral development at around 14 months old. On average, the overall delay between first concerns arising in caregivers and an autism diagnosis is approximately three years. The average age of autism diagnosis has remained unchanged at 4.3 years for decades — and often later for children who are non-white, female, of lower socioeconomic status and in rural areas.

Cognoa’s diagnostic is designed to efficiently diagnose children far sooner, as early as 18 months old, regardless of gender or demographic background. This would enable interventions to begin during an early, critical neurodevelopmental window. Research has shown that children can benefit with twice the cognitive developmental gains should early intervention occur, and by diagnosing children earlier, interventions have the greatest potential to improve lifelong outcomes.

Your tool uses AI and machine learning to help doctors diagnose autism without the need for a behavioral health specialist. Why is early diagnosis so important?

The American Academy of Pediatrics updated their autism guidelines in January 2020, specifically calling for more pediatricians to diagnose autism upon initial presentation in an effort to begin treatment earlier and improve outcomes. Cognoa’s AI is designed to enable pediatricians to do exactly that.

Cognoa’s autism diagnostic harnesses machine learning and AI to allow for the use of large data sets to identify predictive patterns in all children that point toward autism. It is therefore imperative to provide pediatricians with the tools and data that they need to make an accurate diagnosis of autism in the primary care setting, as opposed to referring all children to specialists for evaluation and potential diagnoses that could be achieved in the pediatric office.

In the U.S. today, 1 in 54 children are diagnosed with autism. The reliance on specialists overburdens the health care system, causes undue delays for diagnoses and access to timely interventions and services.

If primary care physicians are better equipped and trained to diagnose autism, specialists will simultaneously benefit. They will be able to focus their time on children with more complex presentations of autism — those who cannot be diagnosed definitively in the primary care setting. This in turn streamlines and accelerates the entire care system, helping pediatricians, specialists, and crucially, children and families living with autism achieve the optimal outcome for the child.

How many people are currently using the tool, and what sort of results has it delivered to date?

Cognoa’s autism diagnostic, which was granted fast-track designation by the FDA, is not currently available to the health care community. We recently submitted our autism diagnostic to the FDA for clearance under priority review on the strength of our successful pivotal trial in which the diagnostic comfortably surpassed all FDA benchmarks. We anticipate that we will introduce our autism diagnostic to the health care and caregiver communities later this year.

Cognoa’s diagnostic is also positioned to be the first FDA-cleared solution that enables diagnosis of autism in the primary care setting.

The goal is to help children receive autism diagnoses earlier in hopes of getting them access to treatment sooner. But there’s still a huge shortage of autism treatment providers. Is Cognoa planning to tackle that issue next?

The greatest obstacle to receiving timely treatment is the diagnostic journey. Without a diagnosis, it is difficult for children and families to access autism-specific interventions, services and benefits.

There are also barriers to treatment once a diagnosis has been made, and similar to the diagnostic journey, the developments in technology and digital solutions provides a unique opportunity to overcome these issues.

At Cognoa, there are several solutions underway in clinical development including a digital therapeutic designed to boost the accessibility of treatment and improve outcomes.

What ultimately needs to happen to fix these pervasive issues in autism diagnosis and treatment long-term?

The primary issue in the diagnostic journey (and thereby the journey to receiving timely interventions with the greatest potential for life changing outcomes) is two-fold: the time it takes from initial concern to a diagnosis and the barriers children and families face to access appropriate care, and the disparity in diagnosing young females and non-white children.

AI and digital solutions are uniquely equipped to aid pediatricians to take a greater role and streamline the overall process, making the overall care journey more efficient for all involved.

The advancements in behavioral health technology provide an enormous opportunity for all families, regardless of demographic background, to access appropriate care sooner and more efficiently. This in turn will benefit the health care community as a whole, who will be more responsive and better able to serve the needs of children and families.

Anything else you’d like to add?

Having been consciously built to embrace gender, racial, ethnic and socio-economic backgrounds, Cognoa’s AI-driven diagnostic also begins to tackle the longstanding issue of disparities in autism diagnoses by reducing unconscious bias from the process.

As it stands, racial and ethnic disparities in autism diagnosis are prevalent with girls diagnosed on average 1.5 years later than boys, and one in four children under the age 8 living with autism, most of them black or Hispanic, not being diagnosed at all. This is due to the lack of access in our current system and because diagnosis historically has been skewed towards autism traits prevalent in white boys, who present characteristics of autism differently than young girls and children who are non-white.

Cognoa’s AI is able to reference troves of external data points and utilize the combined experiences of thousands of physicians, simultaneously assessing many features of a child while analyzing them for predictive autism signals – something a physician (or any human) is simply unable to do. Cognoa’s AI works by analyzing a combination of inputs including: videos of children in natural settings, a parental questionnaire and a health care provider questionnaire.

The AI assesses this data inputted into the diagnostic platform and is then able to pinpoint subtle relationships between different data points, recognizing behavioral patterns that point towards or away from autism. By analyzing such a variety of different human traits and features, unconscious bias is largely removed from the process.